llml 0.2.3

Implementation of basic math data types with high level frontend and low level backend
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83

x86_use!();
use std::ops::{Add, Sub, Mul, Div, Neg};
use crate::{EucVecd2, EucVecf4};

#[derive(Debug)]
#[repr(transparent)]
pub struct EucVecd4 (pub(crate) __m256d);
impl_arith!(EucVecd4, f64, m256);

impl EucVecd4 {
    #[inline(always)]
    pub fn new (x: f64, y: f64, z: f64, w: f64) -> Self {
        unsafe { Self(_mm256_set_pd(w, z, y, x)) }
    }

    #[inline(always)]
    pub fn from_scalar (x: f64) -> Self {
        unsafe { Self(_mm256_set1_pd(x)) }
    }

    #[inline(always)]
    pub fn x (&self) -> f64 {
        unsafe { _mm256_cvtsd_f64(self.0) }
    }

    #[inline(always)]
    pub fn y (&self) -> f64 {
        unsafe { _mm256_cvtsd_f64(_mm256_shuffle_pd(self.0, self.0, _MM_SHUFFLE(1, 1, 1, 1))) }
    }

    #[inline(always)]
    pub fn z (&self) -> f64 {
        unsafe { _mm256_cvtsd_f64(_mm256_shuffle_pd(self.0, self.0, _MM_SHUFFLE(2, 2, 2, 2))) }
    }

    #[inline(always)]
    pub fn w (&self) -> f64 {
        unsafe { _mm256_cvtsd_f64(_mm256_shuffle_pd(self.0, self.0, _MM_SHUFFLE(3, 3, 3, 3))) }
    }

    #[inline(always)]
    pub fn sum (self) -> f64 {
        unsafe {
            let vlow  = _mm256_castpd256_pd128(self.0);
            let vhigh = _mm256_extractf128_pd(self.0, 1); // high 128
            EucVecd2(_mm_add_pd(vlow, vhigh)).sum()
        }
    }

    #[inline(always)]
    pub fn dot (self, rhs: Self) -> f64 {
        (self * rhs).sum()
    }
}


impl PartialEq for EucVecd4 {
    #[inline(always)]
    fn eq (&self, rhs: &Self) -> bool {
        unsafe {
            let cmp = _mm256_cmp_pd(self.0, rhs.0, _MM_CMPINT_EQ);
            let low = *(&_mm256_castpd256_pd128(cmp) as *const __m128d as *const u128);
            let high = *(&_mm256_extractf128_pd(cmp, 1) as *const __m128d as *const u128);
            low == u128::MAX && high == u128::MAX
        }
    }
}

impl Into<EucVecf4> for EucVecd4 {
    #[inline(always)]
    fn into (self) -> EucVecf4 {
        unsafe { EucVecf4(_mm256_cvtpd_ps(self.0)) }
    }
}

impl From<[f64;4]> for EucVecd4 {
    #[inline(always)]
    fn from(x: [f64;4]) -> Self {
        unsafe {
            Self(_mm256_loadu_pd(&x as *const [f64;4] as *const f64))
        }
    }
}